____________________________________________
Did students in the intervention semester have better group project outcomes?
Mann-Whitney U Tests for project grade, perceptions of engagement with the project, and perceptions of project success by condition
Wilcoxon rank sum test with continuity correction
data: average_group_grade by condition
W = 2781, p-value = 0.03567
alternative hypothesis: true location shift is not equal to 0
Wilcoxon rank sum test with continuity correction
data: end_engagement by condition
W = 1712, p-value = 0.02635
alternative hypothesis: true location shift is not equal to 0
Wilcoxon rank sum test with continuity correction
data: success by condition
W = 1432.5, p-value = 0.000152
alternative hypothesis: true location shift is not equal to 0
Did students in the intervention semester view themselves and their peers as better collaborators?
Mann-Whitney U Test P-Values:
General Teammate Evaluation: 0.0079**
Group Accountability: 0.0221*
Individual accountability: 8.613e-11***
NOTE: for the life of me, I can’t get the graph to have different numbers of asterisks on R but I photoshopped it and got it to look right (on Google Doc)
However, students did not like their group mates significantly better in the intervention condition compared to baseline.
Wilcoxon rank sum test with continuity correction
data: like by condition
W = 1980.5, p-value = 0.1633
alternative hypothesis: true location shift is not equal to 0
Did the intervention have an effect on students’ attitudes toward group work in general?
After checking that they were normally distributed, we ran a t-test between final attitude and attitude change during the baseline and intervention semesters. There was not a significant effect of condition on final attitude or attitude change.
Welch Two Sample t-test
data: attitude_change by condition
t = -1.4278, df = 120.75, p-value = 0.1559
alternative hypothesis: true difference in means between group Baseline and group Treatment is not equal to 0
95 percent confidence interval:
-0.33529616 0.05431732
sample estimates:
mean in group Baseline mean in group Treatment
0.02645503 0.16694444
Welch Two Sample t-test
data: final_attitude by condition
t = -0.85513, df = 118.08, p-value = 0.3942
alternative hypothesis: true difference in means between group Baseline and group Treatment is not equal to 0
95 percent confidence interval:
-0.3146278 0.1248490
sample estimates:
mean in group Baseline mean in group Treatment
3.279101 3.373990
___________________________________________________________
We were also curious as to what predicted students’ attitude change. Attitude change was significantly predicted by students’ general teammate evaluation and project grade.
Call:
lm(formula = attitude_change ~ personal_contributions + group_accountability +
individual_accountability + general_teammate_evaluation +
like + success + average_group_grade + condition, data = study_2_clean)
Residuals:
Min 1Q Median 3Q Max
-1.30503 -0.32616 -0.05718 0.33079 1.12671
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.80522 0.67797 -1.188 0.23744
personal_contributions -0.10571 0.08150 -1.297 0.19726
group_accountability -0.03871 0.12957 -0.299 0.76565
individual_accountability -0.03073 0.10515 -0.292 0.77066
general_teammate_evaluation 0.28001 0.10636 2.633 0.00966 **
like 0.06392 0.11405 0.560 0.57629
success 0.03941 0.02155 1.829 0.07006 .
average_group_grade -0.65059 0.28802 -2.259 0.02581 *
conditionTreatment 0.04159 0.12719 0.327 0.74426
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 0.5216 on 113 degrees of freedom
(14 observations deleted due to missingness)
Multiple R-squared: 0.1635, Adjusted R-squared: 0.1043
F-statistic: 2.76 on 8 and 113 DF, p-value: 0.008006
How did students talk about their experience with group work?
LIWC Analyses
The difference at midpoint was not significant.
Thematic Analysis:
Should we use just m1 only like in study 1 or do all midpoint surveys to run intervention-specific hypothesis tests (e.g. students talked more positively ab communication after process reflection) & track experiences across the entire intervention
will also run correlations between themes to get significance
